REVIEWS

Cerebral venous thrombosis

Suzanne M. Silvis1*, Diana Aguiar de Sousa2*, José M. Ferro2 and Jonathan M. Coutinho1
Abstract | Cerebral venous thrombosis (CVT) is an important cause of stroke in young adults. Data
from large international registries published in the past two decades have greatly improved our
knowledge about the epidemiology, clinical manifestations and prognosis of CVT. The
presentation of symptoms is highly variable in this disease, and can range from a patient seen at
the clinic with a 1‑month history of headache, to a comatose patient admitted to the emergency
room. Consequently, the diagnosis of CVT is often delayed or overlooked. A variety of therapies
for CVT are available, and each should be used in the appropriate setting, preferably guided by
data from randomized trials and well-designed cohort studies. Although deaths from CVT have
decreased in the past few decades, mortality remains ~5–10%. In this Review, we provide a
comprehensive and contemporary overview of CVT in adults, with emphasis on advancements
made in the past decade on the epidemiology and treatment of this multifaceted condition.

In 1825, the French physician Ribes reported a case Epidemiology

1
Department of Neurology,
Academic Medical Center,
Meibergdreef 9, 1105 AZ,
Amsterdam, Netherlands.
2
Department of
Neurosciences and Mental
Health (Neurology), Hospital
Santa Maria, University of
Lisbon, Avenida Professor
Egas Moniz, 1649–035,
Lisbon, Portugal.
*These authors contributed
equally to this work
Correspondence to J.M.C.
j.coutinho@amc.nl
doi:10.1038/nrneurol.2017.104
Published online 18 Aug 2017
of a 45‑year-old man who experienced headache and
seizures; autopsy revealed that these symptoms were
caused by thrombosis of the superior sagittal and lat‑
eral sinuses1,2. This case report constitutes the first
detailed description of a patient with cerebral venous
thrombosis (CVT). A few years later, an associ­ation
was found between CVT and pregnancy — or more
precisely, puerperium — for the first time: John
Abercrombie, physician to King George IV, reported
a case of a 24‑year-old woman who, 2 weeks after an
uncomplicated delivery, developed a headache and
multiple seizures3. This patient died as a result of sta‑
tus epilepticus, and autopsy revealed thrombosis of
the superior sagittal sinus and cortical veins. In the
sub­sequent decades, many case reports and small case
series of CVT were produced, but it was not until the
second half of the 20th century, after the introduction
of catheter cerebral angiography, that larger clinical
studies were published, which greatly advanced our
knowledge of the clinical manifestations of and risk
factors for this condition1,4–8. The widespread availabil‑
ity of CT with venography in the late 1980s, and of MRI
with venography several years later, enabled early non-­
invasive diagnosis of CVT9,10. In the past two decades,
findings have been published from large international
registries from all over the world containing data from
hundreds of patients with CVT11–14. In this Review,
we will provide a comprehensive and contemporary
overview of the epidemiology, pathophysiology, diag‑
nosis and treatment of CVT in adults. Paediatric CVT,
which mostly concerns neonates, is beyond the scope of
this Review15.
Incidence. Early estimations of the incidence of CVT
were based on autopsy series16. Extrapolation from an
estimated mortality of 20–50% among patients with
CVT at the time of these series gave an incidence of
0.1–0.2 cases of CVT per 100,000 people. However,
data from population-based studies conducted in
the past few years in the Netherlands and Australia
have shown that the current incidence among adults
is about tenfold higher than this estimate (1.3–1.6
per 100,000)17,18, and the incidence is probably even
higher in Asia and the Middle East, as the rates of
pregnancy and infection-related cases are higher in
these countries19,20. Although the increase in incidence
might partly be explained by a shift in risk factors,
improvements in imaging techniques — which result
in the identification of less-severe cases — is probably
the most important contributing factor.
Risk factors and associated conditions. Most adults with
CVT are aged 20–50 years and <10% of these individuals
are older than 65 years21. Among young and middle-aged
adults, CVT is threefold more common in women than
in men. This heavily skewed sex ratio is the result of the
sex-specific risk factors of oral contraceptives, preg‑
nancy and puerperium22,23. The risk of CVT in women
who use oral contraceptives is increased approximately
sixfold, and this risk is increased further still in women
with obesity who use oral contraceptives24. A large num‑
ber of other risk factors — both transient and permanent
— have been associated with CVT (TABLE 1). Many of
these conditions, such as genetic thrombophilia, inflam‑
matory disorders and cancer, are risk factors for venous

NATURE REVIEWS | NEUROLOGY VOLUME 13 | SEPTEMBER 2017 | 555

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Key points
• Cerebral venous thrombosis (CVT) is an important cause of stroke in young and
middle-aged adults, with a sex ratio heavily skewed towards women
• Manifestations of CVT can be grouped into four distinct clinical syndromes: isolated
intracranial hypertension, focal syndrome, diffuse encephalopathy and cavernous
sinus syndrome
• First-line treatment for CVT is heparin, even in the presence of an intracerebral
haemorrhage
• In a trial completed in 2017, endovascular therapy did not improve the clinical
outcome of patients with severe CVT
• Mortality among patients with CVT has declined in the past few decades to ~8–10%;
although 80% of patients recover without physical disability, many experience
residual chronic symptoms
thrombosis in general. An associ­ation between CVT
and the most common genetic risk factors for thrombo­
philia has been demonstrated in controlled studies25.
Conditions that specifically increase the risk of CVT
include head trauma, arteriovenous malformations,
neurosurgical procedures and infections of the head and
neck26,27. Notably, the prevalence of the risk factors var‑
ies considerably between countries. Dehydration, preg‑
nancy and puerperium, and infections are all prominent
causes of CVT in Asian and Middle Eastern countries,
but are present in fewer than 15% of patients with CVT
in large international and European registries11–13,19,28.
In Mediterranean and Middle Eastern countries, Behçet
disease is an important cause of CVT29. Overall, an
associated condition can be identified in about 85% of
patients11,13.
Pathophysiology
CVT is caused by systemic or local imbalances in pro‑
thrombotic and thrombolytic processes, which lead to
thrombus initiation and propagation in the cerebral
dural sinuses or veins. As venous blood is forced back
into small vessels and capillaries, an increase in venous
and capillary pressure occurs30,31. The specific anatomy of
the brain venous system (FIG. 1), with its extensive anasto‑
moses, often provides sufficient collateral circulation to
compensate for such changes in pressure32,33. However,
when recruitment of collateral pathways does become
insufficient, a disruption of the blood–brain barrier
and decrease in cerebral perfusion pressure develops,
which leads to cerebral oedema, local ischaemia and
often intracerebral haemorrhage30,34,35. Evidence for pro‑
gressive hypoperfusion in experimental models of CVT,
demonstrated by laser Doppler flowmetry and serial
PET ima­ging, further supports the hypothesis that per‑
fusion of the affected brain tissue is still possible in the
initial phases of CVT through collateral drainage path‑
ways32,33. However, a 2015 study could not demonstrate
an association between the extent of baseline intracranial
venous collaterals and the clinical severity or prognosis in
patients with CVT36. Parenchymal lesions in CVT have
frequently been suggested to occur only when the throm‑
bus extends into the cortical veins, but studies in animal
models indicate that occlusion of the major sinuses can
be sufficient to cause venous infarcts33,37. Parenchymal
lesions occur in ~60% of patients with CVT and differ
considerably from those that occur in arterial stroke, as
they cross arterial boundaries, have a haemorrhagic com‑
ponent in almost two-thirds of cases, and often consist of
a combination of vasogenic and cytotoxic oedema11,38–40.
In addition, the dural sinuses play a vital part in
cerebro­spinal fluid absorption. This process is mediated
by the arachnoid villi (also known as Pacchionian gran‑
ulations) that are found in the walls of the sinuses41,42.
Dysfunction of these granulations results in decreased
cerebrospinal fluid absorption and subsequently to
intracranial hypertension43.
Clinical manifestations and diagnosis
Illustrative case history. A 40‑year-old woman was
admitted to the emergency room with a generalized
convulsive seizure. For the previous few days, she had
been complaining of a severe headache. At neurologi‑
cal examination, she opened her eyes on verbal appeal,
made incomprehensible sounds, and localized to pain
with her left arm. In the emergency room, she experi‑
enced four additional generalized convulsive seizures.
CT imaging revealed an intracerebral haemorrhage
(ICH) in the left frontal lobe, with perifocal oedema and
sulcal subarachnoid haemorrhage (FIG. 2a). CT venog‑
raphy revealed thrombosis of the superior sagittal and
left transverse sinus. The patient was started on low-­
molecular-weight heparin (LMWH) and antiepileptic
drugs. After 4 days, her clinical condition deteriorated
and she became comatose. Repeated neuroimaging
showed progressive left-sided oedema and radiological
signs of transtentorial herniation (FIG. 2b), at which point
she underwent emergency, left-sided, decompressive
hemicraniectomy (FIG. 2c). After surgery, she regained
consciousness, but aphasia and hemiparesis remained.
However, after 6 months, she had completely recovered
without any residual functional disabilities.
Clinical manifestations. The symptoms presented
by patients with CVT are highly variable (BOX  1) .
Severe headache is the most common and, usually, the
first symptom of CVT, and is reported by 60–90% of
patients11,12,19,44,45. Some patients report thunderclap
headache that mimics subarachnoid haemorrhage46.
Acute symptomatic seizures — that is, seizures that
occur within 2 weeks of the diagnosis — are present in
30–40% of patients, which is a markedly higher pro‑
portion than seen in the acute phase of arterial stroke
(2–9%) or spontaneous ICH (8–14%)47–53. About 80%
of acute symptomatic seizures actually occur before the
diagnosis has been established54.
Most patients present with a constellation of signs
and symptoms that can be grouped into four distinct
patterns26,55. Patients with isolated intracranial hyper‑
tension experience headache (often accompanied by
nausea), papilloedema, decreased visual acuity, and
tinnitus. Second, patients with thrombosis of the super‑
ficial venous system and parenchymal lesions generally
present with focal neurological deficits, often in com‑
bination with seizures. Thrombosis of the deep venous
system with bilateral oedema of the basal ganglia and

556 | SEPTEMBER 2017 | VOLUME 13 www.nature.com/nrneurol

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

thalami leads to mental status disorder, gaze palsy, dif‑ anaemia, liver disease, kidney disease and inflammatory
fuse encephalopathy, or coma. In rare cases, deep CVT or infectious conditions. The results of genetic testing for
can present with movement disorders56. Finally, throm‑ thrombophilia rarely change management, but this test
bosis of the cavernous sinuses results in orbital pain, can be considered for patients who have no CVT risk
chemosis, ­proptosis and ophthalmoplegia26,55. factors, patients with recurrent thrombosis, patients with
a family history of venous thrombosis, or in the setting
Laboratory investigations. Routine laboratory tests of warfarin-induced skin necrosis25,57,58. Haemoglobin
— including complete blood count, chemistry panel, electrophoresis should be performed in suspected cases
urinalysis, prothrombin time, and activated partial of sickle cell disease or thalassaemia. Lumbar puncture
thromboplastin time — are recommended for all should only be performed in special circumstances, for
patients with CVT57,58. These tests are not helpful to instance when a CNS infection is suspected59.
establish the presence of CVT itself, but they can contrib‑
ute to the identification of associated conditions, such as Imaging. Prompt investigation by noninvasive imaging
is required when CVT is clinically suspected. Magnetic
resonance venography and CT venography are both
Table 1 | Conditions associated with CVT
adequate for diagnosis of CVT, but the former is clearly
Risk Risk factor Prevalence Study type superior for the visualization of brain parenchymal
category in patients lesions55,58,60 (FIG. 3). Theoretically, catheter angio­graphy
with CVT*
remains the most accurate method for diagnosis of CVT
Permanent risk factors but is almost never required anymore. As catheter angi‑
Hereditary Hereditary thrombophilia (total) 34–41% Cohort ography is an invasive technique with a non-negligible
thrombophilia risk of stroke, a patient should only undergo this pro‑
Factor V Leiden thrombophilia 9–13% Case–control
cedure when CT venography or magnetic resonance
Prothrombin Gly20210Ala mutation 9–21% Case–control
venography are inconclusive, a dural arteriovenous fis‑
Antithrombin deficiency 3% Case–control tula is suspected, or when an endovascular therapeutic
Protein S deficiency 2–3% Case–control intervention is planned61–66.
Protein C deficiency 2–5% Case–control The classic sign of acute CVT on unenhanced CT
images is an increased attenuation of the occluded
Systemic Cancer 7% Cohort sinus67,68. Depending on the location of the hyperdense
diseases
Myeloproliferative neoplasms 2–3% Cohort vessel, this finding is sometimes termed a ‘dense triangle
Inflammatory bowel disease 2–3% Cohort sign’ (representing a clot in the superior sagittal sinus) or
Behçet disease 1% Cohort
a ‘cord sign’ (representing thrombosis of cortical or deep
veins)69–71. Contrast injection can reveal an ‘empty delta
Thyroid disease 2% Case reports sign’, which results from contrast enhancement of the
Systemic lupus erythematosus 1% Case series wall of the thrombosed sinus owing to collateral circu‑
Antiphospholipid syndrome 6–17% Cohort lation. However, these signs are only present in a limited
proportion of patients, are even less common in sub­
Nephrotic syndrome 1% Case reports
acute or chronic cases, and are not sufficiently ­specific
Sarcoidosis <1% Case reports for diagnosis of CVT60,68,69,72.
Paroxysmal nocturnal hemiglobinuria NA Case series CT venography provides a detailed depiction of the
Miscellaneous Dural arteriovenous fistula 2% Cohort cerebral venous system, and enables correct identifica‑
tion of sinuses in ~99% of patients and cerebral veins in
Obesity 23% Case–control
~88% of patients73. CT venography is also more sensitive
Transient risk factors to low blood flow than is time‑of‑flight (TOF) magnetic
Sex-specific Oral contraceptives 54–71%‡ Case–control resonance venography, which tends to overestimate the
Pregnancy and puerperium 11–59%‡ Cohort degree of thrombosis in cases of partial vessel occlu‑
sion74–76. However, some anatomic variants can mimic
Hormone replacement therapy 4% ‡
Cohort
CVT, especially sinus atresia, sinus hypoplasia or sinus
Iatrogenic Lumbar puncture 2% Cohort filling defects caused by prominent arachnoid granu‑
Neurosurgical operation 1% Cohort lations. Detection of isolated cortical vein thrombosis
Jugular vein catheterization 1% Cohort can be especially challenging with CT venography65,76.
Nevertheless, CT venography is generally a reliable
Miscellaneous Infections of the head or neck 8–11% Cohort alternative to MRI, particularly in patients with severe
Anaemia 9–27% Case–control CVT in whom MRI is not feasible, or in patients with a
Head trauma 1–3% Cohort contraindication for MRI. Disadvantages of CT venog‑
Spontaneous intracranial hypotension NA Case reports
raphy are exposure to ionizing radiation and the need
for contrast material.
Dehydration 2% Cohort Magnetic resonance venography has several limita‑
*For estimates of the prevalence, we used data from two large cohort studies of patients with tions when used alone, particularly in patients with a
cerebral venous thrombosis (CVT) and data from controlled studies that examined that
particular risk factor11,13. For each risk factor, we list the type of study on which the association hypoplastic sinus, cortical vein thrombosis or partial
is based. ‡Percentage of women. NA, not available. sinus occlusion; consequently, a complete MRI study is

NATURE REVIEWS | NEUROLOGY VOLUME 13 | SEPTEMBER 2017 | 557

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Superior sagittal sinus

Superior anastomotic vein

(vein of Trolard)

Ophthalmic vein
Anterior intercavernous sinus
Inferior sagittal sinus
Sigmoid sinus Sphenoparietal sinus
Transverse sinus
Cavernous sinus
Confluence of sinuses Basal vein of Rosenthal
Superior petrosal sinus

Straight sinus Inferior anastomotic vein

(vein of Labbé)
Vein of Galen Internal cerebral vein
Occipital sinus Internal Jugular vein

Figure 1 | Anatomy of the cerebral venous system. Diagram showing the main components of the cerebral
Nature Reviewsvenous
| Neurology
system. Blue vessels represent the deep venous system.

required to enable diagnosis of CVT60,77. When perform‑
ing MRI, use of T2*-weighted gradient-recalled echo or
susceptibility-weighted imaging is recommended so
as to maximize diagnostic accuracy, especially in the
acute phase of CVT and in patients with suspected iso‑
lated cortical vein thrombosis61. These sequences show
intraluminal thrombi as hypointense areas caused by
increased levels of deoxyhemoglobin65,78–80. Contrast-
enhanced magnetic resonance venography is more
sensitive than TOF magnetic resonance venography,
particularly for sinuses with a small diameter and/or
slow blood flow60,81. Use of 3D T1‑weighted black-blood
sequences has produced promising results in a single
study in patients with subacute CVT82.
When parenchymal lesions are present, characteris‑
tic patterns should raise suspicion of CVT. For example,
venous infarcts often cross arterial boundaries. Clear
demarcation and disproportionate space-occupying
lesions observed with CT in the first hours of onset of
focal signs also suggest CVT58,83. Haemorrhage com‑
monly occurs in combination with oedema, but various
patterns might be observed, ranging from scattered sub‑
cortical foci to a lobar haematoma. Small juxtacortical
haemorrhages were shown to be very specific for throm‑
bosis of the superior sagittal sinus40. Bilateral parenchy‑
mal lesions are found in about one-third of patients with
CVT and nonhaemorrhagic lesions84.
Treatment
General care. A schematic overview of the diagnos‑
tic and therapeutic steps in CVT is provided in FIG. 4.
Current guidelines recommend that patients with
CVT are admitted to a stroke unit58. Any underlying
condition that might have contributed to the disease
should be corrected if possible — especially infection or
dehydration85. Patients with acute symptomatic seizures
should be treated with antiepileptic drugs to prevent
recurrent seiz­ures; those with a supratentorial haemor­
rhagic parenchymal lesion are most at risk of acute
symptomatic seizures47,48,54,86. Prophylactic antiepileptic
drugs could be an option in these patients, although this
approach is not uniformly accepted58,87.
Anticoagulation. Heparin was first introduced as a treat‑
ment for venous thrombosis in the late 1930s. Stansfield,
a British gynaecologist, was one of the first to describe a
favourable outcome for a patient with puerperal CVT
who was treated with heparin. In his 1942 paper, pub‑
lished in the British Medical Journal, Stansfield remarked
that “the introduction of heparin gives us an effective
weapon to treat what has invariably been a fatal com‑
plication of the puerperium, and the clinician’s reward
for an early diagnosis will be the survival of the patient
rather than the sterile pleasure of making an accurate
diagnosis and confirming it in the post-mortem room”
(REF. 88). Stansfield’s decision to use heparin was moti‑
vated by a previous fatal case of puerperal CVT and
encouraging results of heparin treatment in patients with
leg-vein thrombosis.
The use of heparin for treatment of CVT became
more frequent following publication of Stansfield’s paper,
but its use remained controversial for many decades89.
Those who were opposed to heparin therapy were con‑
cerned about the high incidence of ICHs in patients
with CVT11,13,90. On the other hand, advocates of anti­
coagulation argued that withholding heparin could lead

558 | SEPTEMBER 2017 | VOLUME 13 www.nature.com/nrneurol

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

a b c

11mm 3mm

Figure 2 | Imaging findings in a patient with cerebral venous thrombosis. a | Axial noncontrast-enhanced CT scan of an
intracerebral haemorrhage in the left frontal lobe (yellow arrow) with perifocal oedema, as wellNature Reviews
as convexity | Neurology
subarachnoid
haemorrhage (white arrows). Note also the hyperdense aspect of the superior sagittal sinus (red arrow). b | CT scan of the
same patient as in part a, taken a few days later, after she had clinically deteriorated and become comatose. CT scan shows
progressive left-sided oedema (arrow) and radiological signs of transtentorial herniation with a midline shift of
approximately 11 mm. c | Postoperative CT scan of the same patient as in parts a and b after emergency left-sided
decompressive hemicraniectomy was performed, showing a reduction in mass effect, as evidenced by the decrease in
midline shift.

to growth of the thrombus, causing new venous infarcts
and haemorrhages91. In a disease as rare as CVT, clinical
trials to solve this dilemma were a challenge. Nonetheless,
in the 1990s, two small randomized trials were con‑
ducted. A meta-analysis of these trials, which included
a total of 79 patients, showed a nonsignificant difference
in clinical outcome in favour of heparin (relative risk of
death 0.33, 95% CI 0.08–1.21)92–94. Importantly, no new
ICHs occurred in patients treated with heparin, whereas
two patients in the control group had a new ICH, and two
experienced a probable pulmonary embolism. Despite
the small number of patients and statistical uncertainty,
these trials convinced most experts that heparin is bene­
ficial for patients with CVT. Guidelines from both the
European Federation of Neurological Societies and
the American Heart Association now recommend anti­
coagulation with a therapeutic dose of heparin as the pri‑
mary treatment for CVT, regardless of whether patients
have an ICH at baseline58,87.
No consensus exists as to whether unfractionated hep‑
arin (UFH) or LMWH should be used for the treatment
of CVT. In patients with leg-vein thrombosis and pulmo‑
nary embolism, LMWH is associated with a lower risk
of major bleeding, thrombotic complications, and death
than that associated with UFH95. A nonrandom­ized com‑
parison between LMWH and UFH suggests that in CVT,
LMWH also leads to better outcomes96. This observation
has been confirmed by a single-centre randomized trial,
whereas another small trial found no difference in clinical
outcome97,98. The results of both t­ rials must be interpreted
with caution, owing to their low methodological quality.
Surprisingly, despite the rather compelling evidence that
LMWH is superior to UFH for the treatment of CVT,
surveys indicate that many p ­ hysicians — especially
­neurologists — still prefer UFH99,100.
For patients with CVT who are medically stable,
treatment with a vitamin K antagonist should be initi‑
ated. The duration of treatment is generally between 3
and 12 months, or longer in rare cases58,87. The optimal
duration of anticoagulant treatment is not known and
is currently being evaluated in the extending oral anti‑
coagulant treatment after cerebral vein and dural sinus
thrombosis (EXCOA-CVT) study101. The question of
whether patients with CVT can be safely treated with
direct oral anticoagulants (DOACs) has not yet been
answered102,103. In patients with leg-vein thrombosis
or atrial fibrillation, DOACs exhibit a similar efficacy
as heparin followed by vitamin K antagonists, but are
associated with a substantial reduction in the risk of
ICH. The perception that DOACs have a superior safety
profile to that of conventional anticoagulants make
them attractive candidate drugs for the treatment of
CVT104–106, and a phase III trial evaluating the efficacy
and safety of the DOAC dabigatran in patients with CVT
is currently recruiting participants107.
Endovascular treatment. Endovascular treatment for
CVT was first reported in the late 1980s108,109. In the
past two decades, countless case reports and small
case series have been published, but no randomized
trials110,111,112. The success of endovascular treatment for
acute ischaemic stroke has fuelled further enthusiasm
for use of this therapy in CVT; however, importantly,
the transvenous approach that is generally used in
patients with CVT is very different from the endovas‑
cular techniques used in acute ischaemic stroke, despite
the fact that the same thrombectomy devices are some‑
times used. Broadly speaking, two distinct approaches
can be used for the endovascular treatment of CVT.
One is chemical thrombolysis, in which a microcath‑
eter is advanced through the thrombus, and a throm‑
bolytic drug — usually urokinase or recombinant
tissue plasminogen activator — is infused locally112,113.
The other is mechanical thrombectomy with, for
instance, a rheolytic device, balloon angioplasty or a
stent retriever. A combination of both techniques is

NATURE REVIEWS | NEUROLOGY VOLUME 13 | SEPTEMBER 2017 | 559

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Box 1 | Presenting symptoms and signs in CVT
• Headache: 70–90%
• Seizure: 30–40%
• Papilloedema: 30–60%
• Focal motor deficits: 30–50%
• Aphasia: 15–20%
• Mental status disorder: 15–25%
• Coma: 5–15%
• Movement disorder: rare
Data derived from three large cohort studies11,12,45.
sometimes applied. Judging by the number of pub‑
lications, mechanical thrombectomy is increasingly
being used, perhaps because interventionalists have
gained more experience with use of these techniques
in ischaemic stroke112,114–116. Recanalization (partial or
complete) is reported in ~70–90% of patients; how‑
ever, owing to a lack of agreement on how to score
recanalization in CVT, these percentages have limited
value. Reliable data on the risk of complications are
also not available, but systematic reviews of case series
estimate a frequency of postprocedural new ICHs of
10–17%64,110,116. The safety and efficacy of endovascu‑
lar treatment have been assessed in the thrombolysis
or anticoagulation for cerebral venous thrombosis
(TO‑ACT) trial. This trial was terminated prematurely
because of futility. The first results were presented at
the 2017 European Stroke Organization Conference.
In total, 67 patients with severe CVT were randomly
assigned to receive endovascular (65%) or standard
(66%) treatment, and no difference was observed in
the proportion of patients with a good clinical outcome
at 12 months follow‑up. The full results of the TO‑ACT
trial are currently awaited, but on the basis of the avail‑
able information, endovascular treatment should not
be routinely applied in patients with CVT117.
Complications
Hydrocephalus. Some degree of hydrocephalus occurs
in ~15% of patients with CVT14,118. Most of these patients
have obstructive hydrocephalus owing to oedema of the
basal ganglia and thalami, which results from thrombo‑
sis of the deep venous system. Hydrocephalus increases
the risk of a poor clinical outcome, but whether the
increase in risk is due to the hydrocephalus or to the
underlying parenchymal damage is unclear. This uncer‑
tainty and the fact that patients require anticoagulation
means that a shunting procedure should only be consid‑
ered in critically affected patients, in whom no condition
other than the hydrocephalus can explain their clinical
situation118,119.
Intracranial hypertension. Intracranial hypertension
is very common in the acute phase of CVT. In most
patients, symptoms are limited to headache with
or without papilloedema, in which case treatment
can be confined to anticoagulation and analgesics.
Acetazolamide can reduce intracranial pressure to some
extent by lowering cerebrospinal fluid production, but
its effect is insufficient to have a substantial positive
influence in the acute phase of CVT. The use of ster‑
oids has been evaluated in a post-hoc analysis of the
International Study on Cerebral Vein and Dural Sinus
Thrombosis. In this study, steroids were not associated
with an improved outcome. In fact, in patients without
parenchymal lesions, steroids appeared to be detrimen‑
tal. As a result, steroids are not recommended for the
treatment of CVT58,87,120.
Infrequently, patients can develop decreased visual
acuity as a result of intracranial hypertension; in these
individuals, especially in those with acutely threatened
vision, immediate reduction of pressure by a lumbar
puncture or neurosurgical shunting procedure is indi‑
cated58,87,91. In a subset of patients, intracranial hyper‑
tension causes a decrease in consciousness, sometimes
even coma, in the absence of focal parenchymal lesions.
These patients are believed to have severe intracranial
hypertension, resulting in decreased cerebral perfusion.
Evidence-based treatment recommendations cannot be
provided for such cases, as almost no literature exists on
this topic. In rare cases, these patients have been treated
with an emergency shunting procedure or bilateral
decompressive hemicraniectomy119.
Transtentorial herniation. The main cause of early
death in patients with CVT is transtentorial hernia‑
tion owing to mass effect from a parenchymal lesion121.
In  patients with clinical and radiological signs of
impending herniation, decompressive surgery should
be performed 87. A number of studies have shown
that the outcome of patients with CVT and impend‑
ing herni­ation who undergo decompressive surgery
is often favourable122–126. Although these studies were
uncontrolled, we know from previous studies that with‑
out decompressive surgery, most of these patients will
­succumb to the disease127.
Prognosis
The clinical course of CVT is unpredictable in the first
days after diagnosis, and about one quarter of patients
deteriorate in that phase11. Despite the sometimes grim
outlook in the acute phase, multicentre cohort studies
have taught us that the long-term outcome of most
patients with CVT is favourable11. Death in the acute
phase occurs in ~4% of patients and, as mentioned
previously, is generally due to transtentorial herni­
ation11,121,128. Status epilepticus and medical complica‑
tions such as sepsis and pulmonary embolism are other
possible causes of early death in patients with CVT121.
The long-term risk of death is ~8–10%, and about half of
these deaths result from an underlying condition, most
often cancer11,128. A substantial decrease in mortality in
patients with CVT has been observed over the past few
decades129. Although improvement of care and a shift
in risk factors might partly account for this change,
the most important contributing factors are probably
an increased awareness of CVT among clinicians, and
improved imaging techniques, which have led to earlier
diagnosis and detection of less-severe cases.

560 | SEPTEMBER 2017 | VOLUME 13 www.nature.com/nrneurol

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

About 6–10% of surviving patients with CVT have
severe and permanent disability11. Predictors of poor
outcome include older age, male sex, coma, mental
status disorder, ICH, thrombosis of the deep venous
system, infection of the CNS, cancer, and hyperglycae‑
mia at admission11,130,131. However, an infection outside
of the CNS is not associated with a poor outcome in
Left temporal parenchymal lesion
a b
Bilateral thalamic oedema
c d
patients with CVT28. Excellent outcomes can gener‑
ally be expected for patients who present with isolated
intracranial hypertension11,132. On the other hand, mor‑
tality of >30% has been reported in series of patients
with severe CVT admitted to an intensive care unit133,134.
Despite the fact that ~80% of patients recover from
CVT without physical disability, many of these patients
do experience residual chronic symptoms. About half of
patients report headache during follow‑up, and severe
headaches that require bed rest or hospital admission
persist in 14% of patients11,135. More than half of survivors
of CVT report subtle neuropsychological difficulties or
depression. These complaints are often associated with a
negative effect on employment status: ~20–40% of patients
are unable to return to their prior working life135–138. In
one study, a low level of education was associated with an
increased risk of unemployment after CVT136.
Remote seizures occur in ~10% of patients after CVT,
a far lower proportion than that seen in the acute phase
of CVT. Remote seizures are more likely to occur in
patients with early seizures, motor deficits, and supra­
tentorial lesions — especially if haemorrhagic47,136 Severe
visual loss due to intracranial hypertension is rare, but
ophthalmological evaluation should be performed in
patients with papilloedema or visual complaints139.

Juxtacortical haemorrhage
e scarce. Persistent intracranial hypertension or chronic
Figure 3 | Illustrative brain parenchymal lesions in patients with cerebral venous
thrombosis. a | Fluid attenuation inversion recovery (FLAIR) Nature
shows aReviews | Neurology
left temporal
parenchymal lesion (arrow) in a patient with thrombosis of the left lateral sinus and
jugular vein. b | Magnetic resonance venography in the same patient as in part a shows
an absence of flow in the left lateral sinus and jugular vein (arrows). c | FLAIR shows
bilateral oedema of the thalami and basal ganglia in a patient with thrombosis of the
deep venous system (arrows). d | Magnetic resonance venography in the same patient as
in part c shows no visible filling of the deep venous system (arrows). e | Axial noncontrast-
enhanced CT scan of a small juxtacortical haemorrhage with surrounding oedema (white
arrow) in a patient with superior sagittal thrombosis and cortical vein thrombosis. Note
the hyperdense cortical veins, indicating thrombosis (yellow arrows).
Recanalization
The rate of spontaneous venous recanalization in
patients with CVT is ~85%128. Analysis of the time to
recanalization indicates that this process predom‑
inantly occurs in the first few months after CVT, but
that it can take up to 1 year140–142. Studies of the relation‑
ship between recanalization and clinical outcome have
produced conflicting results128,140,141,143. Most of these
studies have only examined an association between
disability (measured by the modified Rankin scale)
and recanalization, so data regarding the association
of ­recanalization with other long-term complaints are
headache after CVT might indicate the existence of a
post-thrombotic syndrome, similar to that seen in some
patients with leg-vein thrombosis141. Whether persis‑
tent occlusion increases the risk of CVT recurrence also
remains a poorly addressed issue. Such an association
has been shown in paediatric patients but compara‑
ble data for adults are not available141,144. Despite the
scant data on recanalization and clinical outcome, docu‑
mentation of the extent of recanalization with follow‑up
imaging can be useful in clinical practice as it facilitates
diagnostic work‑up in patients for whom recurrence of
CVT is suspected. Without routine follow‑up imaging,
determining whether thrombosis has recurred when a
patient experiences a new-onset h ­ eadache after CVT is
often very difficult58.
Recurrence and long-term management
The risk of a new cerebral or systemic venous thrombotic
event after an episode of CVT is ~4 per 100 person–
years, and most recurrences are within the first year145.
Male sex and polycythaemia or thrombocythaemia
are established risk factors for recurrence145,146. Severe

NATURE REVIEWS | NEUROLOGY VOLUME 13 | SEPTEMBER 2017 | 561

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

Clinical suspicion of CVT

• Severe headache (60–90% of patients)
• Acute symptomatic seizures (30–40% of patients)
• Manifestations of CVT can be grouped into four main clinical
syndromes: isolated intracranial hypertension, focal syndrome,
diffuse encephalopathy and cavernous sinus syndrome

Routine laboratory investigation Imaging*

• Complete blood count Magnetic resonance
• Chemistry panel venography or CT
• Urinalysis venography
• Prothrombin and activated
partial thromboplastin time

No
Confirmed CVT? Consider other diagnoses

Yes
Decompressive hemicraniectomy
Yes Decompressive surgery should be performed in
Impending transtentorial herniation? patients with clinical and radiological signs of
impending herniation.
No

Initiate low-molecular- Search for underlying

weight heparin treatment‡
• Low-molecular-weight
heparin usually preferred
over unfractionated heparin
• Intracranial haemorrhage
due to CVT is not a
contraindication for heparin
treatment
aetiology
• Associated risk factors include
hereditary thrombophilia,
systemic disease (e.g. cancer),
sex-specific factors and
iatrogenic factors
• Screening for hereditary
thrombophilia can be
considered, but rarely changes
management

Management of complications
Yes • Hydrocephalus
Complications? • Intracranial hypertension
• Transtentorial herniation
No

Initiate oral anticoagulant treatment when clinically stable

Dependent on the underlying aetiology, duration of
anticoagulant treatment is generally 3–12 months.

Figure 4 | Schematic overview of diagnostic and therapeutic steps in CVT. *MRI is the most suitable modality for
detecting brain parenchymal lesions; magnetic resonance venography and CT venography areNature equallyReviews
adequate | Neurology
for diagnosis of cerebral venous thrombosis (CVT), but magnetic resonance venography is superior for detecting cortical
vein thrombosis and establishing the age of the thrombus; CT venography is sufficient in the acute phase in patients who
are severely affected or have a contraindication for MRI. ‡The use of low-molecular-weight heparin is generally preferred
over unfractionated heparin, except when the need for rapid reversal of anticoagulation is anticipated, for instance
because of emergency neurosurgical intervention.

thrombophilia and previous noncerebral venous throm‑
boembolism have also been associated with an increased
risk of recurrence in some cohorts13,147. However, results
regarding the association between recurrence and dura‑
tion of anticoagulation are conflicting13,145,147, and a trial
comparing short-term (3–6 months) with long-term
(12 months) a­ nticoagulation is currently ongoing101.
Oestrogen-containing contraception and hormo‑
nal therapy increase the risk of thrombosis and, con‑
sequently, women should be advised not to use these
drugs after CVT. Although pregnancy is associated
with an increased risk of venous thrombotic events, the
absolute risk of recurrent events related to subsequent
pregnancy among women who have a history of CVT
is low: estimated rates are nine cases of CVT and 27 of
noncerebral venous thromboembolism per 1,000 preg‑
nancies11,58,148,149. A history of CVT, therefore, should not
be a contraindication for future pregnancies, but women
of child-bearing age should be informed of the increased
relative risk of pregnancy-related thrombotic events and
the possible benefit of antithrombotic prophylaxis149.
Future research directions
Burning clinical issues that remain to be addressed
include the efficacy and safety of endovascular treat‑
ment, the optimal duration of anticoagulant treatment,
and whether DOACs can be safely applied in these
patients. The TO‑ACT, EXCOA-CVT and RE‑SPECT

562 | SEPTEMBER 2017 | VOLUME 13 www.nature.com/nrneurol

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

CVT trials, respectively, are currently addressing these Conclusions

three issues, and results are expected in the next few
years101,107,117. The prospective DECOMPRESS‑2 regis‑
try, in which CVT patients who undergo decompressive
surgery are included, will provide an improved forecast
of the clinical outcome in these patients. An increased
understanding of the risk factors that are associated
with CVT, both genetic and acquired, is also required.
Efforts have been made to form an international collab‑
oration to identify new genes and biomarkers associated
with CVT150.
In the past few years, we have seen major advancements
in our knowledge of the epidemiology, diagnosis and
treatment of CVT. Once considered to be an invari­
ably fatal condition, CVT is now viewed as a disease
with a generally favourable prognosis. Although more
common than previously believed, CVT is still a fairly
infrequent disease. As a result, our only hope to gain
further insights into the genetics, associated conditions
and treatment of this multifaceted condition is through
international collaborations.

1. Kalbag, R. M. & Woolf, A. L. Cerebral Venous
Thrombosis (Oxford Univ. Press, 1967).
2. Ribes, M. F. Des recherches faites sur la phlebite.
Revue Medicale Française et Etrangère et Journal
de Clinique de l’Hôtel-Dieu et de la Charité de Paris. 3,
5–41 (1825).
3. Abercrombie, J. Pathological and Practical
Researches on Diseases of the Brain and Spinal Cord
(Waugh and Innes, 1828).
4. Huhn, A. Clinical aspects of intracranial venous
thrombosis [German]. Radiologe 11, 377–390
(1971).
5. Krayenbuhl, H. A. Cerebral venous and sinus
thrombosis. Clin. Neurosurg. 14, 1–24 (1966).
6. Bansal, B. C., Gupta, R. R. & Prakash, C. Stroke during
pregnancy and puerperium in young females below
the age of 40 years as a result of cerebral venous/
venous sinus thrombosis. Jpn. Heart J. 21, 171–183
(1980).
7. Bousser, M. G., Chiras, J., Bories, J. & Castaigne, P.
Cerebral venous thrombosis — a review of 38 cases.
Stroke 16, 199–213 (1985).
8. Bousser, M. G. Cerebral venous thrombosis. Report of
76 cases [French]. J. Mal. Vasc. 16, 249–254 (1991).
9. Goldberg, A. L. et al. Computed tomography of dural
sinus thrombosis. J. Comput. Assist. Tomogr. 10,
16–20 (1986).
10. Mattle, H. P. et al. Cerebral venography with MR.
Radiology 178, 453–458 (1991).
11. Ferro, J. M. et al. Prognosis of cerebral vein and dural
sinus thrombosis: results of the International Study on
Cerebral Vein and Dural Sinus Thrombosis (ISCVT).
Stroke 35, 664–670 (2004).
12. Pai, N., Ghosh, K. & Shetty, S. Hereditary
thrombophilia in cerebral venous thrombosis: a study
from India. Blood Coagul. Fibrinolysis 24, 540–543
(2013).
13. Dentali, F. et al. Long-term outcomes of patients with
cerebral vein thrombosis: a multicenter study.
J. Thromb. Haemost. 10, 1297–1302 (2012).
14. Wasay, M. et al. Cerebral venous thrombosis: analysis
of a multicenter cohort from the United States.
J. Stroke Cerebrovasc. Dis. 17, 49–54 (2008).
15. Roach, E. S. et al. Management of stroke in infants
and children: a scientific statement from a Special
Writing Group of the American Heart Association
Stroke Council and the Council on Cardiovascular
Disease in the Young. Stroke 39, 2644–2691 (2008).
16. Stam, J. Cerebral venous and sinus thrombosis:
incidence and causes. Adv. Neurol. 92, 225–232
(2003).
17. Devasagayam, S., Wyatt, B., Leyden, J. & Kleinig, T.
Cerebral venous sinus thrombosis incidence is higher
than previously thought: a retrospective population-
based study. Stroke 47, 2180–2182 (2016).
18. Coutinho, J. M., Zuurbier, S. M., Aramideh, M. &
Stam, J. The incidence of cerebral venous thrombosis:
a cross-sectional study. Stroke 43, 3375–3377
(2012).
19. Khealani, B. A. et al. Cerebral venous thrombosis:
a descriptive multicenter study of patients in Pakistan
and Middle East. Stroke 39, 2707–2711 (2008).
20. Janghorbani, M. et al. Cerebral vein and dural sinus
thrombosis in adults in Isfahan, Iran: frequency and
seasonal variation. Acta Neurol. Scand. 117, 117–121
(2008).
21. Ferro, J. M. et al. Cerebral vein and dural sinus
thrombosis in elderly patients. Stroke 36, 1927–1932
(2005).
22. de Bruijn, S. F., Stam, J., Koopman, M. M. &
Vandenbroucke, J. P. Case–control study of risk of
cerebral sinus thrombosis in oral contraceptive users
who are carriers of hereditary prothrombotic
conditions. The Cerebral Venous Sinus Thrombosis
Study Group. BMJ 316, 589–592 (1998).
23. Cantu, C. & Barinagarrementeria, F. Cerebral
venous thrombosis associated with pregnancy
and puerperium. Review of 67 cases. Stroke 24,
1880–1884 (1993).
24. Zuurbier, S. M. et al. Risk of cerebral venous
thrombosis in obese women. JAMA Neurol. 73,
579–584 (2016).
25. Lauw, M. N., Barco, S., Coutinho, J. M. &
Middeldorp, S. Cerebral venous thrombosis and
thrombophilia: a systematic review and meta-analysis.
Semin. Thromb. Hemost. 39, 913–927 (2013).
26. Stam, J. Thrombosis of the cerebral veins and sinuses.
N. Engl. J. Med. 352, 1791–1798 (2005).
27. Coutinho, J. M. Cerebral venous thrombosis.
J. Thromb. Haemost. 13 (Suppl. 1), S238–S244
(2015).
28. Zuurbier, S. M. et al. Clinical outcome of anticoagulant
treatment in head or neck infection-associated
cerebral venous thrombosis. Stroke 47, 1271–1277
(2016).
29. Yesilot, N. et al. Cerebral venous thrombosis in
Behcet’s disease compared to those associated with
other etiologies. J. Neurol. 256, 1134–1142 (2009).
30. Ueda, K., Nakase, H., Miyamoto, K., Otsuka, H. &
Sakaki, T. Impact of anatomical difference of the
cerebral venous system on microcirculation in a
gerbil superior sagittal sinus occlusion model.
Acta Neurochir. (Wien) 142, 75–82 (2000).
31. Gotoh, M., Ohmoto, T. & Kuyama, H. Experimental
study of venous circulatory disturbance by dural sinus
occlusion. Acta Neurochir. (Wien) 124, 120–126
(1993).
32. Ungersbock, K., Heimann, A. & Kempski, O. Cerebral
blood flow alterations in a rat model of cerebral sinus
thrombosis. Stroke 24, 563–569 (1993).
33. Schaller, B. et al. Hemodynamic changes after
occlusion of the posterior superior sagittal sinus:
an experimental PET study in cats. AJNR Am.
J. Neuroradiol. 24, 1876–1880 (2003).
34. Rottger, C. et al. A new model of reversible sinus
sagittalis superior thrombosis in the rat: magnetic
resonance imaging changes. Neurosurgery 57,
573–580 (2005).
35. Schaller, B. & Graf, R. Cerebral venous infarction: the
pathophysiological concept. Cerebrovasc. Dis. 18,
179–188 (2004).
36. Barboza, M. A., Mejias, C., Colin-Luna, J., Quiroz-
Compean, A. & Arauz, A. Intracranial venous
collaterals in cerebral venous thrombosis: clinical and
imaging impact. J. Neurol. Neurosurg. Psychiatry 86,
1314–1318 (2015).
37. Kurokawa, Y., Hashi, K., Okuyama, T. & Uede, T.
Regional ischemia in cerebral venous hypertension
due to embolic occlusion of the superior sagittal sinus
in the rat. Surg. Neurol. 34, 390–395 (1990).
38. Lovblad, K. O. et al. Diffusion-weighted MR in cerebral
venous thrombosis. Cerebrovasc. Dis. 11, 169–176
(2001).
39. Mullins, M. E., Grant, P. E., Wang, B., Gonzalez, R. G.
& Schaefer, P. W. Parenchymal abnormalities
associated with cerebral venous sinus thrombosis:
assessment with diffusion-weighted MR imaging.
AJNR Am. J. Neuroradiol. 25, 1666–1675 (2004).
40. Coutinho, J. M. et al. Small juxtacortical hemorrhages
in cerebral venous thrombosis. Ann. Neurol. 75,
908–916 (2014).
41. Pollay, M. The function and structure of the
cerebrospinal fluid outflow system. Cerebrospinal
Fluid Res. 7, 9 (2010).
42. Brunori, A., Vagnozzi, R. & Giuffre, R. Antonio
Pacchioni (1665–1726): early studies of the dura
mater. J. Neurosurg. 78, 515–518 (1993).
43. Leach, J. L., Meyer, K., Jones, B. V. & Tomsick, T. A.
Large arachnoid granulations involving the dorsal
superior sagittal sinus: findings on MR imaging and
MR venography. AJNR Am. J. Neuroradiol. 29,
1335–1339 (2008).
44. Coutinho, J. M. et al. Cerebral venous thrombosis in
the absence of headache. Stroke 46, 245–247
(2015).
45. Ferro, J. M. et al. Cerebral vein and dural sinus
thrombosis in Portugal: 1980–1998. Cerebrovasc.
Dis. 11, 177–182 (2001).
46. de Bruijn, S. F., Stam, J. & Kappelle, L. J. Thunderclap
headache as first symptom of cerebral venous sinus
thrombosis. CVST Study Group. Lancet 348,
1623–1625 (1996).
47. Ferro, J. M. et al. Seizures in cerebral vein and dural
sinus thrombosis. Cerebrovasc. Dis. 15, 78–83
(2003).
48. Davoudi, V., Keyhanian, K. & Saadatnia, M. Risk
factors for remote seizure development in patients
with cerebral vein and dural sinus thrombosis. Seizure
23, 135–139 (2014).
49. Kalita, J., Chandra, S. & Misra, U. K. Significance of
seizure in cerebral venous sinus thrombosis. Seizure
21, 639–642 (2012).
50. Lamy, C. et al. Early and late seizures after cryptogenic
ischemic stroke in young adults. Neurology 60,
400–404 (2003).
51. Bladin, C. F. et al. Seizures after stroke: a prospective
multicenter study. Arch. Neurol. 57, 1617–1622
(2000).
52. Passero, S., Rocchi, R., Rossi, S., Ulivelli, M. & Vatti, G.
Seizures after spontaneous supratentorial
intracerebral hemorrhage. Epilepsia 43, 1175–1180
(2002).
53. De Herdt, V. et al. Early seizures in intracerebral
hemorrhage: incidence, associated factors, and
outcome. Neurology 77, 1794–1800 (2011).
54. Ferro, J. M. et al. Early seizures in cerebral vein and
dural sinus thrombosis: risk factors and role of
antiepileptics. Stroke 39, 1152–1158 (2008).
55. Bousser, M. G. & Ferro, J. M. Cerebral venous
thrombosis: an update. Lancet Neurol. 6, 162–170
(2007).
56. Kalita, J., Bhoi, S. K., Chandra, S. & Misra, U. K.
Reversible parkinsonian features in deep cerebral
venous sinus thrombosis. Can. J. Neurol. Sci. 40,
740–742 (2013).
57. Bushnell, C. et al. Guidelines for the prevention of
stroke in women: a statement for healthcare
professionals from the American Heart Association/
American Stroke Association. Stroke 45, 1545–1588
(2014).
58. Saposnik, G. et al. Diagnosis and management of
cerebral venous thrombosis: a statement for
healthcare professionals from the American Heart
Association/American Stroke Association. Stroke 42,
1158–1192 (2011).
59. Canhao, P. et al. Safety of lumbar puncture in patients
with cerebral venous thrombosis. Eur. J. Neurol. 20,
1075–1080 (2013).
60. Leach, J. L., Fortuna, R. B., Jones, B. V. & Gaskill-
Shipley, M. F. Imaging of cerebral venous thrombosis:
current techniques, spectrum of findings, and
diagnostic pitfalls. Radiographics 26 (Suppl. 1),
S19–S41 (2006).
61. Lafitte, F. et al. MRI and MRA for diagnosis and
follow‑up of cerebral venous thrombosis (CVT).
Clin. Radiol. 52, 672–679 (1997).

NATURE REVIEWS | NEUROLOGY VOLUME 13 | SEPTEMBER 2017 | 563

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
REVIEWS

62. Liang, L. et al. Evaluation of the intracranial dural
sinuses with a 3D contrast-enhanced MP‑RAGE
sequence: prospective comparison with 2D‑TOF MR
venography and digital subtraction angiography.
AJNR Am. J. Neuroradiol. 22, 481–492 (2001).
63. Qu, H. & Yang, M. Early imaging characteristics of 62
cases of cerebral venous sinus thrombosis. Exp. Ther.
Med. 5, 233–236 (2013).
64. Siddiqui, F. M. et al. Mechanical thrombectomy in
cerebral venous thrombosis: systematic review of 185
cases. Stroke 46, 1263–1268 (2015).
65. Coutinho, J. M., Gerritsma, J. J., Zuurbier, S. M. &
Stam, J. Isolated cortical vein thrombosis: systematic
review of case reports and case series. Stroke 45,
1836–1838 (2014).
66. Conforto, A. B. et al. Dural arteriovenous fistula and
cerebral venous thrombosis. Arq. Neuropsiquiatr. 73,
548 (2015).
67. Roland, T. et al. Unenhanced brain CT is useful to
decide on further imaging in suspected venous sinus
thrombosis. Clin. Radiol. 65, 34–39 (2010).
68. Buyck, P. J. et al. CT density measurement and H:H
ratio are useful in diagnosing acute cerebral venous
sinus thrombosis. AJNR Am. J. Neuroradiol. 34,
1568–1572 (2013).
69. Virapongse, C., Cazenave, C., Quisling, R., Sarwar, M.
& Hunter, S. The empty delta sign: frequency and
significance in 76 cases of dural sinus thrombosis.
Radiology 162, 779–785 (1987).
70. Ford, K. & Sarwar, M. Computed tomography of
dural sinus thrombosis. AJNR Am. J. Neuroradiol 2,
539–543 (1981).
71. Rizzo, L. et al. Cerebral venous thrombosis: role of CT,
MRI and MRA in the emergency setting. Radiol. Med.
115, 313–325 (2010).
72. Black, D. F., Rad, A. E., Gray, L. A., Campeau, N. G. &
Kallmes, D. F. Cerebral venous sinus density on
noncontrast CT correlates with hematocrit. AJNR Am.
J. Neuroradiol. 32, 1354–1357 (2011).
73. Linn, J. et al. Diagnostic value of multidetector-row CT
angiography in the evaluation of thrombosis of the
cerebral venous sinuses. AJNR Am. J. Neuroradiol.
28, 946–952 (2007).
74. Ozsvath, R. R. et al. Cerebral venography: comparison
of CT and MR projection venography. AJR Am.
J. Roentgenol. 169, 1699–1707 (1997).
75. Wetzel, S. G. et al. Cerebral veins: comparative study
of CT venography with intraarterial digital subtraction
angiography. AJNR Am. J. Neuroradiol. 20, 249–255
(1999).
76. Rodallec, M. H. et al. Cerebral venous thrombosis and
multidetector CT angiography: tips and tricks.
Radiographics 26 (Suppl. 1), S5–S18 (2006).
77. Ayanzen, R. H. et al. Cerebral MR venography: normal
anatomy and potential diagnostic pitfalls. AJNR Am.
J. Neuroradiol. 21, 74–78 (2000).
78. Idbaih, A. et al. MRI of clot in cerebral venous
thrombosis: high diagnostic value of susceptibility-
weighted images. Stroke 37, 991–995 (2006).
79. Boukobza, M., Crassard, I., Bousser, M. G. &
Chabriat, H. MR imaging features of isolated cortical
vein thrombosis: diagnosis and follow‑up. AJNR Am.
J. Neuroradiol. 30, 344–348 (2009).
80. Robinson, R. J. & Bhuta, S. Susceptibility-weighted
imaging of the brain: current utility and potential
applications. J. Neuroimaging 21, e189–e204 (2011).
81. Farb, R. I. et al. Intracranial venous system:
gadolinium-enhanced three-dimensional MR
venography with auto-triggered elliptic centric-ordered
sequence — initial experience. Radiology 226,
203–209 (2003).
82. Niu, P. P. et al. Diagnosis of non-acute cerebral venous
thrombosis with 3D T1‑weighted black blood
sequence at 3T. J. Neurol. Sci. 367, 46–50 (2016).
83. Bakac, G. & Wardlaw, J. M. Problems in the diagnosis
of intracranial venous infarction. Neuroradiology 39,
566–570 (1997).
84. Ferro, J. M. et al. Cerebral venous thrombosis with
nonhemorrhagic lesions: clinical correlates and
prognosis. Cerebrovasc. Dis. 29, 440–445 (2010).
85. Bushnell, C. & Saposnik, G. Evaluation and
management of cerebral venous thrombosis.
Continuum (Minneap. Minn.) 20, 335–351 (2014).
86. Masuhr, F. et al. Risk and predictors of early epileptic
seizures in acute cerebral venous and sinus
thrombosis. Eur. J. Neurol. 13, 852–856 (2006).
87. Einhaupl, K. et al. EFNS guideline on the treatment
of cerebral venous and sinus thrombosis in adult
patients. Eur. J. Neurol. 17, 1229–1235 (2010).
88. Stansfield, F. R. Puerperal cerebral thrombophlebitis
treated by heparin. Br. Med. J. 1, 436–438 (1942).
89. Krayenbuhl, H. A. Cerebral venous and sinus
thrombosis. Neurol. Med. Chir. (Tokyo) 10, 1–24
(1968).
90. Gettelfinger, D. M. & Kokmen, E. Superior sagittal
sinus thrombosis. Arch. Neurol. 34, 2–6 (1977).
91. Coutinho, J. M. & Stam, J. How to treat cerebral
venous and sinus thrombosis. J. Thromb. Haemost. 8,
877–883 (2010).
92. Einhaupl, K. M. et al. Heparin treatment in sinus
venous thrombosis. Lancet 338, 597–600 (1991).
93. de Bruijn, S. F. & Stam, J. Randomized, placebo-
controlled trial of anticoagulant treatment with low-
molecular-weight heparin for cerebral sinus
thrombosis. Stroke 30, 484–488 (1999).
94. Coutinho, J., de Bruijn, S. F., Deveber, G. & Stam, J.
Anticoagulation for cerebral venous sinus thrombosis.
Cochrane Database Syst. Rev. 8, CD002005 (2011).
95. Erkens, P. M. & Prins, M. H. Fixed dose subcutaneous
low molecular weight heparins versus adjusted dose
unfractionated heparin for venous thromboembolism.
Cochrane Database Syst. Rev. 9, CD001100 (2010).
96. Coutinho, J. M. et al. Unfractionated or low-molecular
weight heparin for the treatment of cerebral venous
thrombosis. Stroke 41, 2575–2580 (2010).
97. Misra, U. K., Kalita, J., Chandra, S., Kumar, B. &
Bansal, V. Low molecular weight heparin versus
unfractionated heparin in cerebral venous sinus
thrombosis: a randomized controlled trial.
Eur. J. Neurol. 19, 1030–1036 (2012).
 venous thrombosis? Stroke 39, 105–110 (2008).
98. Afshari, D. et al. The efficacy and safety of low-
molecular-weight heparin and unfractionated heparin
in the treatment of cerebral venous sinus thrombosis.
Neurosciences (Riyadh) 20, 357–361 (2015).
99. Field, T. S., Camden, M. C., Al‑Shimemeri, S., Lui, G. &
Lee, A. Y. Antithrombotic strategy in cerebral venous
thrombosis: differences between neurologist and
hematologist respondents in a Canadian survey.
Can. J. Neurol. Sci. 44, 116–119 (2017).
100. Coutinho, J. M. et al. Treatment variations in cerebral
venous thrombosis: an international survey.
Cerebrovasc. Dis. 32, 298–300 (2011).
101. ISRCTN registry. EXCOA-CVT study: the benefit of
EXtending oral antiCOAgulation treatment after acute
Cerebral Vein Thrombosis. ISRCTN http://www.isrctn.
com/ISRCTN25644448 (2014).
102. Mendonca, M. D., Barbosa, R., Cruz‑e‑Silva, V.,
Calado, S. & Viana-Baptista, M. Oral direct thrombin
inhibitor as an alternative in the management of
cerebral venous thrombosis: a series of 15 patients.
Int. J. Stroke 10, 1115–1118 (2015).
103. Geisbusch, C., Richter, D., Herweh, C., Ringleb, P. A.
& Nagel, S. Novel factor xa inhibitor for the treatment
of cerebral venous and sinus thrombosis: first
experience in 7 patients. Stroke 45, 2469–2471
(2014).
104. Robertson, L., Kesteven, P. & McCaslin, J. E.
Oral direct thrombin inhibitors or oral factor Xa
inhibitors for the treatment of pulmonary embolism.
Cochrane Database Syst. Rev. 12, CD010957
(2015).
105. Robertson, L., Kesteven, P. & McCaslin, J. E. Oral
direct thrombin inhibitors or oral factor Xa inhibitors
for the treatment of deep vein thrombosis.
Cochrane Database Syst. Rev. 6, CD010956 (2015).
106. Chatterjee, S., Sardar, P., Biondi-Zoccai, G. &
Kumbhani, D. J. New oral anticoagulants and the
risk of intracranial hemorrhage: traditional and
Bayesian meta-analysis and mixed treatment
comparison of randomized trials of new oral
anticoagulants in atrial fibrillation. JAMA Neurol. 70,
1486–1490 (2013).
107. US National Library of Medicine. ClinicalTrials.gov
https://clinicaltrials.gov/ct2/show/NCT02913326
(2017).
108. Higashida, R. T., Helmer, E., Halbach, V. V. &
Hieshima, G. B. Direct thrombolytic therapy for
superior sagittal sinus thrombosis. AJNR Am.
J. Neuroradiol 10, S4–S6 (1989).
109. Scott, J. A., Pascuzzi, R. M., Hall, P. V. & Becker, G. J.
Treatment of dural sinus thrombosis with local
urokinase infusion. Case report. J. Neurosurg. 68,
284–287 (1988).
110. Canhao, P., Falcao, F. & Ferro, J. M. Thrombolytics
for cerebral sinus thrombosis: a systematic review.
Cerebrovasc. Dis. 15, 159–166 (2003).
111. Ciccone, A., Canhao, P., Falcao, F., Ferro, J. M. &
Sterzi, R. Thrombolysis for cerebral vein and dural
sinus thrombosis. Cochrane Database Syst. Rev. 1,
CD003693 (2004).
112. Stam, J., Majoie, C. B., van Delden, O. M.,
van Lienden, K. P. & Reekers, J. A. Endovascular
thrombectomy and thrombolysis for severe cerebral
sinus thrombosis: a prospective study. Stroke 39,
1487–1490 (2008).
113. Frey, J. L., Muro, G. J., McDougall, C. G., Dean, B. L. &
Jahnke, H. K. Cerebral venous thrombosis: combined
intrathrombus rtPA and intravenous heparin. Stroke
30, 489–494 (1999).
114. Li, G. et al. Safety and validity of mechanical
thrombectomy and thrombolysis on severe cerebral
venous sinus thrombosis. Neurosurgery 72, 730–738
(2013).
115. Caso, V., Billeci, A. M. & Leys, D. Interventional
neuroradiology in the treatment of cerebral venous
thrombosis. Front. Neurol. Neurosci. 23, 144–160
(2008).
116. Borhani Haghighi, A. et al. Mechanical thrombectomy
for cerebral venous sinus thrombosis: a comprehensive
literature review. Clin. Appl. Thromb. Hemost. 20,
507–515 (2014).
117. Coutinho, J. M. et al. Thrombolysis or anticoagulation
for cerebral venous thrombosis: rationale and design
of the TO‑ACT trial. Int. J. Stroke 8, 135–140 (2013).
118. Zuurbier, S. M. et al. Hydrocephalus in cerebral venous
thrombosis. J. Neurol. 262, 931–937 (2015).
119. Lobo, S. et al. Shunting in acute cerebral venous
thrombosis: a systematic review. Cerebrovasc. Dis. 37,
38–42 (2014).
120. Canhao, P. et al. Are steroids useful to treat cerebral
121. Canhao, P. et al. Causes and predictors of death in
cerebral venous thrombosis. Stroke 36, 1720–1725
(2005).
122. Ferro, J. M. et al. Decompressive surgery in
cerebrovenous thrombosis: a multicenter registry and
a systematic review of individual patient data. Stroke
42, 2825–2831 (2011).
123. Aaron, S. et al. Decompressive craniectomy in cerebral
venous thrombosis: a single centre experience.
J. Neurol. Neurosurg. Psychiatry 84, 995–1000
(2013).
124. Zuurbier, S. M. et al. Decompressive hemicraniectomy
in severe cerebral venous thrombosis: a prospective
case series. J. Neurol. 259, 1099–1105 (2012).
125. Zhang, S., Zhao, H., Li, H., You, C. & Hui, X.
Decompressive craniectomy in hemorrhagic cerebral
venous thrombosis: clinicoradiological features and
risk factors. J. Neurosurg. http://dx.doi.org/
10.3171/2016.8.JNS161112 (2016).
126. Mohindra, S., Umredkar, A., Singla, N., Bal, A. &
Gupta, S. K. Decompressive craniectomy for malignant
cerebral oedema of cortical venous thrombosis:
an analysis of 13 patients. Br. J. Neurosurg. 25,
422–429 (2011).
127. Theaudin, M. et al. Should decompressive surgery be
performed in malignant cerebral venous thrombosis?:
a series of 12 patients. Stroke 41, 727–731 (2010).
128. Dentali, F., Gianni, M., Crowther, M. A. & Ageno, W.
Natural history of cerebral vein thrombosis: a
systematic review. Blood 108, 1129–1134 (2006).
129. Coutinho, J. M., Zuurbier, S. M. & Stam, J. Declining
mortality in cerebral venous thrombosis: a systematic
review. Stroke 45, 1338–1341 (2014).
130. Ferro, J. M. et al. Risk score to predict the outcome of
patients with cerebral vein and dural sinus thrombosis.
Cerebrovasc. Dis. 28, 39–44 (2009).
131. Zuurbier, S. M. et al. Admission hyperglycemia and
clinical outcome in cerebral venous thrombosis. Stroke
47, 390–396 (2016).
132. Biousse, V., Ameri, A. & Bousser, M. G. Isolated
intracranial hypertension as the only sign of cerebral
venous thrombosis. Neurology 53, 1537–1542
(1999).
133. Kowoll, C. M. et al. Severe cerebral venous and sinus
thrombosis: clinical course, imaging correlates, and
prognosis. Neurocrit. Care 25, 392–399 (2016).
134. Soyer, B. et al. Outcome of a cohort of severe cerebral
venous thrombosis in intensive care. Ann. Intensive
Care 6, 29 (2016).
135. Koopman, K. et al. Long-term sequelae after cerebral
venous thrombosis in functionally independent
patients. J. Stroke Cerebrovasc. Dis. 18, 198–202
(2009).
136. Hiltunen, S., Putaala, J., Haapaniemi, E. &
Tatlisumak, T. Long-term outcome after cerebral
venous thrombosis: analysis of functional and
vocational outcome, residual symptoms, and adverse
events in 161 patients. J. Neurol. 263, 477–484
(2016).
137. Bugnicourt, J. M. et al. Cognitive impairment after
cerebral venous thrombosis: a two-center study.
J. Neurol. 260, 1324–1331 (2013).

564 | SEPTEMBER 2017 | VOLUME 13 www.nature.com/nrneurol

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.
 REVIEWS

138. Buccino, G., Scoditti, U., Patteri, I., Bertolino, C. &
Mancia, D. Neurological and cognitive long-term
outcome in patients with cerebral venous sinus
thrombosis. Acta Neurol. Scand. 107, 330–335
(2003).
139. Purvin, V. A., Trobe, J. D. & Kosmorsky, G. Neuro-
ophthalmic features of cerebral venous obstruction.
Arch. Neurol. 52, 880–885 (1995).
140. Herweh, C. et al. Frequency and temporal profile of
recanalization after cerebral vein and sinus
thrombosis. Eur. J. Neurol. 23, 681–687 (2016).
141. Arauz, A. et al. Time to recanalisation in patients with
cerebral venous thrombosis under anticoagulation
therapy. J. Neurol. Neurosurg. Psychiatry 87,
247–251 (2016).
142. Putaala, J., Hiltunen, S., Salonen, O., Kaste, M. &
Tatlisumak, T. Recanalization and its correlation to
outcome after cerebral venous thrombosis. J. Neurol.
Sci. 292, 11–15 (2010).
143. Strupp, M., Covi, M., Seelos, K., Dichgans, M. &
Brandt, T. Cerebral venous thrombosis: correlation
between recanalization and clinical outcome — a long-
term follow‑up of 40 patients. J. Neurol. 249,
1123–1124 (2002).
144. Kenet, G. et al. Risk factors for recurrent venous
thromboembolism in the European collaborative
paediatric database on cerebral venous thrombosis:
a multicentre cohort study. Lancet Neurol. 6,
595–603 (2007).
145. Miranda, B. et al. Venous thromboembolic events after
cerebral vein thrombosis. Stroke 41, 1901–1906
(2010).
146. Lim, H. Y., Ng, C., Donnan, G., Nandurkar, H. & Ho, P.
Ten years of cerebral venous thrombosis: male gender
and myeloproliferative neoplasm is associated with
thrombotic recurrence in unprovoked events.
J. Thromb. Thrombolysis 42, 423–431 (2016).
147. Martinelli, I. et al. Long-term evaluation of the risk
of recurrence after cerebral sinus-venous thrombosis.
Circulation 121, 2740–2746 (2010).
148. Ruiz-Sandoval, J. L. et al. Cerebral venous
thrombosis in a Mexican multicenter registry of
acute cerebrovascular disease: the RENAMEVASC
study. J. Stroke Cerebrovasc. Dis. 21, 395–400
(2012).
149. Aguiar de Sousa, D., Canhao, P. & Ferro, J. M. Safety
of pregnancy after cerebral venous thrombosis:
a systematic review. Stroke 47, 713–718 (2016).
150. Cotlarciuc, I. et al. Towards the genetic basis of
cerebral venous thrombosis — the BEAST
Consortium: a study protocol. BMJ Open 6, e012351
(2016).
Author contributions
S.M.S. and D.A.d.S. researched the data for the article and
co‑wrote the first draft. J.M.F. and J.M.C. made substantial
contribution to discussion of the context an edited the manu­
script before submission. J.M.C. supervised the preparation
of the manuscript.
Competing interests statement
J.M.F. has received personal fees from Boehringer Ingelheim
and Daiichi Sankyo. J.M.C. has received research grants for
CVT from two non-profit organizations: the Dutch Thrombosis
Society and the Netherlands Brain Foundation, and is a steer-
ing committee member of the RE-SPECT CVT trial, a clinical
trial that evaluates the efficacy and safety of dabigatran for
the treatment of CVT sponsored by Boehringer Ingelheim .
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

NATURE REVIEWS | NEUROLOGY VOLUME 13 | SEPTEMBER 2017 | 565

©
2
0
1
7
M
a
c
m
i
l
l
a
n
P
u
b
l
i
s
h
e
r
s
L
i
m
i
t
e
d
,
p
a
r
t
o
f
S
p
r
i
n
g
e
r
N
a
t
u
r
e
.
A
l
l
r
i
g
h
t
s
r
e
s
e
r
v
e
d
.